Human mucin (MUC) family member, MUC1, is a transmembrane glycoprotein expressed in low levels by normal, primarily, epithelial cells and high levels by cancer cells. Most epithelial adenocarcinomas, including those from the breast and ovary, over-express MUC1 and, more importantly, may promote anti-MUC1 antibodies, which correlate with more favorable prognosis. By extension, we hypothesized that the risk for cancer might be reduced by pre-existing MUC1-specific immunity and assembled persuasive cross-sectional and case-control data, which suggest that various medical and reproductive events, including breast mastitis, IUD use, tubal ligation, and bone fracture may predict the presence of anti-MUC1 antibodies and lower ovarian cancer risk. To validate these "retrospective" observations, we now propose a prospective study with three aims. Under aim 1, we will identify women either before or at the earliest stages of events that may affect MUC1 immunity and document changes in humoral and cellular immune responses related to MUC1. Under aim two, we will collect (pre-operative) blood specimens from women with ovarian cancer prior to any therapy to identify similarities or differences between tumor-generated immunity (both humoral and cellular) and the immunity that may be generated by the medical and reproductive events studied in aim 1. A goal of these analyses will be to identify the epitopes of MUC1 that are immunogenic in this setting and the types of immune responses generated in order that we may refine the assays used for measuring anti- MUC1 antibodies in aim 3. Under this aim, we will perform a nested case-control study within the Nurses Health Study (NHS) to assess directly the predictive value of anti-MUC1 antibodies many years prior to the development of ovarian cancer. We will select NHS cohort members with a plasma specimen available at least 4 years before they developed ovarian cancer and matched controls and measure anti-MUC1 antibodies to determine actual, rather than inferred, predictive value of antibodies using prospective epidemiologic data. The findings of our study could provide the basis for developing a safe cancer- preventing vaccine for ovarian cancer and perhaps other MUC1-expressing tumors.